For White Girls, a Bigger Penalty for Being Obese

Photo: Tobyotter

We hear increasingly about the healthcare costs of obesity; but what about social costs?

A forthcoming Economics and Human Biology paper (abstract here; PDF here) by Mir Ali, Aliaksandr Amialchuk, and John Rizzo, titled “The Influence of Body Weight on Social Network Ties Among Adolescents,” makes this interesting argument:

We find that obese adolescents have fewer friends and are less socially integrated than their non-obese counterparts. We also find that such penalties in friendship networks are present among whites but not African-Americans or Hispanics, with the largest effect among white females.

As Ali added in an e-mail:

What we find is that it’s mostly White adolescents who are marginalized for being heavier and the effect is the strongest among overweight White females — i.e. few people are likely to be friends with these overweight White females and if they do have friends, those friends are also more likely to be on the fringe of the social network. There’s no social marginalization penalty for overweight African Americans or Hispanics. We hypothesize that this is mainly due to racial differences in ideal body weight norms – a thinner body ideal for Whites compared to African-Americans or Hispanics.

In related (and more far-reaching) obesity research, another paper in Economics and Human Biology paper (abstract here; PDF here) challenges the conventional wisdom on the relationship between poverty and obesity. From “Overweight and Poor?” by Dean Jolliffe of the World Bank:

Contrary to conventional wisdom, NHANES data indicate that the poor have never had a statistically significant higher prevalence of overweight status at any time in the last 35 years. Despite this empirical evidence, the view that the poor are less healthy in terms of excess accumulation of fat persists. This paper provides evidence that conventional wisdom is reflecting important differences in the relationship between income and the body mass index. The first finding is based on distribution-sensitive measures of overweight which indicates that the severity of overweight has been higher for the poor than the nonpoor throughout the last 35 years. The second finding is from a newly introduced estimator, unconditional quantile regression (UQR), which provides a measure of the income-gradient in BMI at different points on the unconditional BMI distribution. The UQR estimator indicates that the strongest relationship between income and BMI is observed at the tails of the distribution. There is a statistically significant negative income gradient in BMI at the obesity threshold and some evidence of a positive gradient at the underweight threshold. Both of these UQR estimates imply that for those at the tails of the BMI distribution, increases in income are correlated with healthier BMI values.

And, since good things come in three, here’s one more interesting recent obesity paper, from Charles Courtemanche, Garth Heutel, and Patrick McAlvanah, called “Impatience, Incentives, and Obesity” (abstract here; PDF here):

This paper explores the relationship between time preferences, economic incentives, and body mass index (BMI). Using data from the 2006 National Longitudinal Survey of Youth, we first show that greater impatience increases BMI and the likelihood of obesity even after controlling for demographic, human capital, occupational, and financial characteristics as well as risk preference. Next, we provide evidence of an interaction effect between time preference and food prices, with cheaper food leading to the largest weight gains among those exhibiting the most impatience. The interaction of changing economic incentives with heterogeneous discounting may help explain why increases in BMI have been concentrated amongst the right tail of the distribution, where the health consequences are especially severe. Lastly, we model time-inconsistent preferences by computing individuals’ quasi-hyperbolic discounting parameters (beta and delta). Both long-run patience (delta) and present-bias (beta) predict BMI, suggesting obesity is partly attributable to rational intertemporal tradeoffs but also partly to time inconsistency.

So, to summarize today’s news in obesity: white girls are disproportionately punished for obesity; a poor person is not necessarily more likely to be obese than a non-poor person (although the severity may be greater); and you are less likely to be obese if you are more patient.

Calories burned while reading this too-long blog post: roughly 150.

Carry on.


I have to ask the burning question... What is that in the picture? A very stiff rope? A piece of rebar?


Interesting stuff. I wonder might the reverse process also be plausible - that socially excluded girls might end up comfort-eating and gaining weight? (Don't have time to read the paper at the moment, sorry if that has already been covered!)


This is based on survey data, so it's entirely possible. Also, obese individuals tend to cluster into groups, which would drive the same type of correlation. However, the paper was trying to prove causality and isolate these effects.

Hexe Froschbein

You could also view the way people react to obese girls as an instant test of character for acquaintances -- would you really want to be friends with someone who is that kind of person?

So, on the whole I think this is a good thing to know, and now that you know about this trick to ferret out unfriendly folk -- if you're going to a party to meet new potential friends, take a fat girl along for maximum efficiency 8():

John B

"There’s no social marginalization penalty for overweight African Americans or Hispanics. We hypothesize that this is mainly due to racial differences in ideal body weight norms – a thinner body ideal for Whites compared to African-Americans or Hispanics."

I question how the author makes that conclusion. Perhaps, as a white researcher (and dealing with other white researchers), seeing heavier blacks and hispanics bothers him less than seeing heavy white girls.

Also, the % of black girls being heavy is significantly higher so that would reduce the social marginalization.

in shape overweight black girl

Personally, as a black woman and a researcher, I am interested in knowing more about this also. I was unable to get the pdf, however, as with most research, there are always limitations. Yes, there are differences in what is perceived as "fat" or "thin" among various cultures, however, there are always those who are considered outliers. I am one of those outliers. I would, however, attribute my ideals of fat and thin to the fact that I grew up in Iowa. Maybe it has more to do with environment, rather than a particular cultural background? Perhaps the reason why social marginalization is not among the blacks and hispanics (as much- this really depends on who you would talk to) because they don't except the Eurocentrically built BMI measurement system?


White women are the least likely of the six (white, black, or Hispanic; male or female) to be obese. It makes sense to me that the more of a minority you are, in the sense of obesity, the greater social costs you pay.


actually, the reverse seems true as well- that skinny black women are picked on by their peers more than fat ones- i personally know black women who want to gain weight for aesthetic reasons, and have seen one skinny one nicknamed "olive oil" (noting the irony in this context of calling her white)


Oh, the plight of the white.

This is almost entirely a function of a white, Eurocentric standard of beauty. Skinny white girls are the ideal. Obese white girls violate this and, as a result, are most ostracized. Black girls, skinny or otherwise, simply can't measure up from the get go.

robin marlowe

Ya know, the question of who qualifies as "obese" has not been answered here. I know a kid who might by your standards and she has more real friends than I can count on both hands. They are all close and look out for each other. On the other hand, I myself might qualify and have very few real close friends. Have not taken the time to cultivate them, I have been told by that one. She is correct? So, now what? AT least, when I get rid of the weight on my back, I can say finished, my way! I once thought that i had a few, and they both let me down. Neither one acted like a friend. Perhaps my standards are just too high. Honestly, I do not know the answer to that question. But it would be nice to have a few more.


I also know of many slender women, most of them white over the years who've told me they've been picked on and degendered, i.e "you look like a boy", it would be uncomfortable to have sex with you (seriously) because they are seen as too thin. Some have complained that its harder to put on weight than it is too lose it and all the help is for the fat nobody ever remembers the thin, etc., I've found you can't even predict by looking who's going to feel this way as some of them did not look thin enough to my eyes to feel too 'underweight' but there you go.

I also wonder how much of a class element, would working class white people be more like ethnic minority communities or like middle and upper class white people where this hatred of fat (and love of thin) is much more of a marker of class identity, hence the poor people are fatter assumption. My experience would suggest WCW would share more values with ethnic minorities for similar reasons (woman are valued differently).


Fiona Mackenzie

Those women are not complaining, they are bragging. They are calling attention to how slender they are. Skinny women do it all the time.


The unique thing is we are the only country where are poor are heavier than our rich. The point is that we make it simple for the poor to get the less healthy food, in abundance. Where the rich can get healthy foods that are priced higher. The fat white girl probably has less means, where as the fat black girl is more racially accepted. No matter what her means are.

Disco Stu

This is something our society needs to change. All overweight people should be marginalized equally, not just whites.

Mika K

This is a very interesting study, but slightly misleading in its findings regarding poverty and overweight - what they seem to be saying is that the population distribution (so your likelihood) of overweight is the same regardless of your income, but if you are poor and overweight, you are much more likely to be very overweight. I would be interested to see the data disaggregated for age and disability / incapacity - being poor you are statistically likely to die younger (from ill-health) and suffer from disability-affected life years than your richer friends. Therefore a comparison across age bands would be interesting to see how this affects the statistical significance of being poor and overweight.